Increasingly, clinical and non-human primate studies suggest that dysfunction in identified neural networks linking the limbic structures and the prefrontal cortex results in behavioral and emotional patterns that resemble autism. Neuroimaging techniques to date have not consistently identified brain abnormalities associated with autism. This likely due to heterogeneity of autistic populations, inadequate control populations and limitations of previous imaging techniques themselves. This project will use technological advances in structural and spectroscopic magnetic resonance imaging (sMRI and 1/H-MRSI and 1/H- MRSI) to investigate abnormalities in identified limbic-prefrontal neural networks and determine their association with autism. Seventy-two children and adolescents with autism aged 7;0 to 18;11 years and a group comparable in age, gender and IQ without autism will receive high resolution sMRI. From the MR images, total and separate brain tissue (white matter/gray matter, cerebrospinal fluid) volume measurements will be performed (a) on the different structures of the limbic regions, i.e. the amygdala and hippocampus and (b) on two regions of the prefrontal cortex, i.e., the orbitofrontal and dorsolateral white/gray matter volume differences in orbitofrontal cortex of all autistic people as compared to control subjects; that volumetric differences in the structures of the M-ORB/AMYG circuit will correlate more strongly with neuropsychological test indices measuring functions of the amygdala and the orbitofrontal cortex, and clinical and behavioral measures of autistic symptoms (e.g. severity of autism), than with IQ measures; and that the dorsolateral prefrontal-hippocampal (DL/HIPPO) circuit will be more severely affected in low-functioning children and adolescents with autism than in high-functioning children and adolescents with autism and controls. In addition, a subgroup of subjects including 20 young high- functioning children and adolescents persons with autism, age 11; 0 to 18; 11 years and 20 controls matched for age, gender, IQ, and handedness will receive 1/H-MRSI to assess the chemical composition for identified cerebral areas. In addition, all monkeys with infant and adult lesions in defined cerebral structures will be behaviorally tested (Project III) and imaged using the same technique to identify brain regions affected by the early versus late damage to the amygdala or orbitofrontal cortex. We hypothesize that abnormalities seen in higher-functioning children and adolescents with autism will be localized in the M-ORB/AMYG circuit, and those seen in the DL/HIPPO circuit will be associated with intellectual impairment; abnormalities of NAA reflecting alterations in neuronal integrity of the M-ORB-AMYG in autistic people will more strongly correlate with clinical and behavioral traits of autism as well as with scores on neuropsychological tasks measuring function of the M- ORB/AMYG circuit (Project I); in monkeys, we hypothesize that an early (infantile) lesion within the M-ORB/AMYG circuit (Project III) will have functional ramifications in other neural circuits (DL/HIPPO) and that the same lesion performed in adulthood will not have the same widespread impact.